Last Week in AIAnthropicWednesday, January 21, 2026

LWiAI Podcast #231 - Claude Cowork, Anthropic $10B, Deep Delta Learning

AI-Generated Summary

This episode covers Anthropic's major $10 billion funding round and discusses Claude Cowork, a collaborative AI feature that enables multiple users to work together within Claude's interface. The funding milestone signals investor confidence in Anthropic's technical capabilities and competitive position against OpenAI, while Claude Cowork represents a significant product shift toward team-based AI workflows that could reshape how organizations deploy Claude in their operations.

Key Takeaways

This episode covers Anthropic's major $10 billion funding round and discusses Claude Cowork, a collaborative AI feature that enables multiple users to work together within Claude's interface.
The funding milestone signals investor confidence in Anthropic's technical capabilities and competitive position against OpenAI, while Claude Cowork represents a significant product shift toward team-based AI workflows that could reshape how organizations deploy Claude in their operations.

Read the full article on Last Week in AI

Read on Last Week in AI

TechCrunch6h ago

Anthropic temporarily banned OpenClaw’s creator from accessing Claude

Anthropic

Anthropic revoked access to its Claude AI model for the creator of OpenClaw, a tool designed to bypass Claude's safety guidelines and enable jailbreaking. This escalation reflects the intensifying conflict between AI safety measures and adversarial testing, signaling that major AI labs are willing to take punitive action against those who publicly distribute circumvention tools—a critical moment for how the industry enforces acceptable use policies.

Wired6h ago

Anthropic’s Mythos Will Force a Cybersecurity Reckoning—Just Not the One You Think

Anthropic

Anthropic's new Mythos AI model is generating significant concern in cybersecurity circles, with many viewing it as a potentially powerful tool for malicious hackers. The model's capabilities have prompted industry experts to sound alarms about vulnerabilities that could be exploited at scale. However, security researchers emphasize that the real issue isn't the model itself, but rather systemic weaknesses that have persisted in software development practices.

The Register6h ago

Project Glasswing and open source software: The good, the bad, and the ugly

Anthropic

Anthropic has announced Project Glasswing, a coalition of major technology companies pledging $100 million in AI resources to identify and remediate previously undiscovered vulnerabilities in critical open source software. The initiative leverages artificial intelligence to systematically scan widely-used open source projects for security flaws that may have gone undetected for years.

The Register6h ago

Anthropic will let your agents sleep on its couch

Anthropic

Anthropic has introduced Managed Agents, a new service designed to help organizations deploy AI agents that can handle ongoing business tasks with minimal human intervention. The service aims to simplify the process of creating and maintaining autonomous AI systems that can operate continuously on behalf of companies.

TechCrunch6h ago

Is Anthropic limiting the release of Mythos to protect the internet — or Anthropic?

Anthropic

Anthropic announced this week that it is restricting the release of its newest AI model, Mythos, citing concerns that the system is too capable of identifying security exploits in widely-used software. The company framed the limitation as a responsible safety measure to protect internet infrastructure and users globally from potential harm if the model's capabilities were freely available.